TEXAS REDISTRICTING & ELECTION LAW
Here’s an update on where things stand, as of noon today, on the election law bills in the Texas Legislature.
It’s a beautiful thing when something complex can be made a bit simpler through data visualization.
A higher res version and some additional notes here.
The first known county-level electoral map from the 1880 presidential contest:
An even more fun, zoomable version available here.
[Note: The modern red/blue distinction is inverted on this map, i.e., red = Democrats, blue = Republicans]
So how did turnout rates in Texas in 2012 vary by gender?
In general, Texas mirrored national trends, with Texas women outvoting Texas men by a little over three points.
However, since there are more women than men in Texas as elsewhere - and since women are registered at a slightly higher rate than men (69.2% vs. 64.5%) - Texas women’s higher turnout rate meant that women ended up making up 55.25% of the universe of actual Texas voters vs. 44.75% for men.
Perhaps predictably, there were some variances by age.
While women outvoted men in all but one age group, women in the younger age groups proved especially more responsible than their male counterparts (even though their turnout rates trailed older women by a significant margin).
That at least appears to be Texas Attorney General Greg Abbott is hinting at - though the ultimate call will be Governor Perry’s.
Christy Hoppe at the Dallas Morning News has more.
PAC+ and the One Texas PAC yesterday released the toplines of extensive polling they are conducting on Texas Hispanic voters in Dallas, Harris, Bexar, Tarrant, and El Paso counties.
The poll had a margin of error of +/- 1.9% and was conducted using both cellphones and landlines.
Here are some of the tidbits:
- 69% of Texas Hispanic voters went to primary or secondary school in Texas.
- 81% have lived at their current address for at least 5 years.
- 21% believe Senator Ted Cruz is a Democrat. Another 18% were unsure.
- 11% of respondents said they were not registered to vote (when in fact they were). Another 3% said they were not sure if they were registered. Of the 11% who said they weren’t registered, 51% of them said they had been registered at one point - suggesting either they didn’t understand the way voter registration works or maybe that they moved and needed to update their registration.
When it comes to a preference for Hispanic candidates, a plurality said they were at least somewhat more likely to vote for a Hispanic candidate, with a quarter saying they were much more likely:
But voters in the survey also were not particularly politically aware, with only 47% correctly identifying the next gubernatorial election as being in 2014:
And, as for the billion dollar question, why don’t people vote?
The full toplines here.
The Census Bureau’s estimate last week that African-Americans outvoted Anglos in 2012 for the first time - ever - has gotten a lot of buzz.
But it also has drawn some skeptical questions.
An earlier piece on this blog suggested that the Texas data pointed to at least some measure of exaggeration by Texas African-Americans about having voted - perhaps predictable given the presence of Barack Obama on the ballot.
And this morning, Pew is out with a piece that looks at all 50 states plus the District of Columbia and finds the presence of a noticeable “social desirability bias” skewing African-American turnout rates across the nation - in some cases by double digits.
That’s a numerator problem - in other words, the inflation of turnout rates by overstating the number of people who voted.
Interestingly, Pew found the overreporting of voting was more pronounced in 2008 and 2012 than in 2004 and speculated:
Might this be because non-voting blacks were more eager than non-voting whites to tell survey takers that they voted for the first ever African-American president? While there’s no way of knowing for sure, the data are suggestive. When we plotted the state discrepancies in 2008 and 2004, we found a similar pattern, but we also found the racial skew was stronger in 2008 and 2012, the two elections in which Obama was on the ballot, than in 2004.
But that’s not quite the end of the story.
Others have pointed out that there also may be a denominator problem - again, especially with African-American voters.
That problem arises because the Census Bureau assumes all citizens of voting age are eligible to vote. But with African-American men, in particular, that isn’t true because of felony records. That problem works the other direction - artificially understating turnout rates by including people who can’t vote. For African-American men, the impact is more than 8 points.
On the other hand, simply looking at the voter roll isn’t dispositive because many states, like Texas, don’t track the ethnicity of voters, making it necessary to model the voter file.
All of which is simply to say what seems to be a straightforward enough question is full of nuances and caveats.
Updated chart here reflecting action through Wednesday.
This afternoon, the Texas House voted down a proposed constitutional amendment, SJR 13, that would have imposed a two-term limit on statewide officeholders (other than judges).
The vote was 61-80 and was not on party lines.
The amendment had cleared the Texas Senate earlier this session by a vote of 27-4.
Early voting is more popular than ever in Texas. But who is voting early?
If you guessed younger voters (more hip to electronic machines, etc.), you would be in for a surprise.
That’s because 2012 data from the 15 largest Texas counties shows exactly the opposite. The older you get, the more likely you are to vote early.
Now, the data shown in the chart includes mail ballots but those are cast primarily by people over 65 and the chart shows a clear inverse age correlation well before most people are eligible to cast mail ballots.
Whether the trend is driven by things like having busy schedules as you get older - or by ethnicity (younger voters in Texas are much more heavily non-Anglo and there could be cultural preferences at play) or some other factor - is not possible to say on this data. But it is an interesting side note nonetheless.
An earlier post took a look at the Census Bureau’s estimates of voter turnout in Texas by age - and, well, the stats weren’t pretty for 18-24 voters compared with their peers in other states.
But can the voter file tell us more?
The good news is that the age of voters - unlike ethnicity - is something that voter registrars in Texas track, so the data is considerably more certain and available. (To be sure, there are a few thousand voters without recorded birthdates and some inevitable coding errors - but not enough to be statistically significant in a universe of 13.6 million registered voters.)
So - drum roll - let’s take a look at the actual turnout data, starting with the voter pool.
Heading into the last general election, 18-24 year olds made up 9.61% of all registered voters in Texas, while voters 24-34 made up another 17.6% of the voter pool.
Voters over 50 made up 46.2% of the pool of potential voters.
But by this point, it will come as no surprise that when it came to actual turnout, older voters dominated.
In fact, voters over 50 - a group that is disproportionately Anglo - made up 55.3% of all actual voters.
By contrast, voters under 35 made up just 18.7% of actual voters, with voters 18-24 being just 6.3% of the voter universe.
It’s not all bad news, however. Looking at statewide turnout rates as a percentage of registered voters, the data looks a little better for those young voters who actually got past the point of getting registered.
While 25.4% of all eligible 18-24 year olds voted in 2012, of those who were registered 38.13% did.
But that figure disguises some marked regional differences.
Among the 15 largest counties, for example, 46.29% of young voters in Travis County went to the polls compared with 69% of Travis County voters over 65.
By contrast, only 26.34% of their peers in Cameron County and 31.19% of those in El Paso County.
And, oh, how many 18-24 year olds were registered?
Using the Census Bureau’s estimate of citizen voting age population, it looks like about 56% applying numbers from actual voting records.
The Census Bureau, though, records a lower rate base on surveys (38.3%), likely meaning that many young people may not be aware they registered (e.g. registered at the DMV office and forgot) or may have moved since they first registered.
Earlier posts (here and here) looked at estimates of voter turnout in Texas by ethnicity.
But how does that breakdown at the county level?
The Census Bureau’s data doesn’t get that granular, but, using estimates from an analysis of the Texas voter file (caveats explained here), the results for big 5 counties look broadly consistent with state-level estimates - with a couple of exceptions.
The biggest variance is significantly stronger performance among African-Americans and Hispanics in Tarrant County when compared both with other counties’ and with statewide turnout rates. In fact, African-Americans in Tarrant County came less than two points away from topping the Anglo turnout rate.
A large part of the credit no doubt goes to State Sen. Wendy Davis, who bested former State Rep. Mark Shelton in the state’s most expensive non-federal race.
The second notable exception is stronger performance among Hispanics (and Asians) in Travis County.
Flipping to look at just Hispanic performance, the results continued to vary fairly dramatically across the state:
Although initially there was excitement among some observers about the strength of early voting in parts of the Valley and border region counties, the early burst appears to have been more about people shifting to early voting rather than more people voting.
After some spirited debate, the Texas House today gave preliminary approval 99-46 95-50 to SB 346, Sen. Seliger’s bill to require 501(c)(4) groups and other politically active non-profits to disclose their donors once the group’s political spending crosses a $25,000 calendar-year threshold.
The bill has drawn criticism from various interest groups but has been especially controversial with several conservative groups.
The bill gets its second and final vote tomorrow, and, if it survives unamended, goes to Gov. Perry for signature.
On the Senate side, HB 148, which would place limitations on the number of mail ballots a person could mail for another, moved out of the Senate state affairs committee today on a 5-3 vote.
The chart of election-law bills has been updated to reflect the new status of the bills as well as the handful of other bills voted out of committee today or scheduled for floor votes this week. As before, new information is listed in red on the chart.
On Sunday - a day after school board elections failed to elect a Hispanic member - Hispanic groups in the Grand Prairie ISD filed suit against the district in federal court.
Contending that section 2 of the Voting Rights Act requires the district to move from at-large to single-member districts, the suit called the current system “pernicious” and said that the system was “intended to safeguard the political power of the Anglo majority and, thus, to deny representation to the GPISD’s citizens of voting age who are Hispanic.”
The complaint noted that “33.8 percent of the citizen voting age population is Hispanic and 63.5 percent of students are Hispanic” and said that “since 1999, bloc voting by Anglos has resulted in the defeat of all other Hispanic candidates (at least four).”
The case has been assigned to US District Judge Sidney Fitzwater.
A copy of the complaint can be found here.
